Article Text

Download PDFPDF
Cohort study
Raised blood glucose as a predictor of dementia risk in adults with and without diabetes
  1. Leigh Martin Riby1,
  2. Deborah Michelle Riby2
  1. 1Department of Psychology, Northumbria University, Newcastle, UK
  2. 2Department of Psychology, Durham University, Durham, UK
  1. Correspondence to : Dr Leigh Martin Riby, Department of Psychology, Northumbria University, City Campus, Newcastle NE1 8ST, UK; leigh.riby{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Commentary on: OpenUrlCrossRefPubMedWeb of Science


Research has suggested that diabetic status doubles the risk of dementia.1 There is also a suggestion that efficient glucose regulation in healthy older adults contributes to the stability of cognitive skills, particularly in domains subserved by the hippocampus (eg, episodic memory which is known to decline in dementia).2 Consequently, the present work is a timely contribution in understanding the relationship between the blood glucose and risk for dementia.


The incidence of dementia was assessed longitudinally at 2-year intervals for 2067 dementia-free older adults from the Adult Changes in Thought Study. This community cohort consists of 839 men, 1228 women, 232 with diabetes at baseline, and a baseline mean age of 76 years. The cognitive abilities screening instrument, which tested attention, short-term and long-term memory, visuospatial ability and executive function, was used alongside clinical reports, imaging and neuropsychological data to confirm dementia status. A composite measure of glucose regulatory status was gathered at baseline and throughout the study, assessing levels of glycosylated haemoglobin, fasting glucose and random glucose. The authors also collated evidence for other risk factors, including apolipoprotein E, lifestyle, cerebrovascular disease and hypertension.


Cox-regression models examined the relationship between glucose levels and the incidence of dementia stratified by diabetic and cerebrovascular disease status, and controlling for: age at study entry, study cohort, sex, educational level, exercise level, blood pressure, coronary artery disease, atrial fibrillation, smoking and treatment for hypertension. Baseline values of glucose for the diabetic and non-diabetic samples were 175 and 101 mg/dL, respectively. After an average follow-up period of 6.8 years, dementia is developed in 26.1% (n=450/1724) of the patients with non-diabetes and in 21.6% (n=74/343) of the diabetes. For the full sample the paper reported 19.5% with probable or possible Alzheimer's disease (AD), 2.7% vascular dementia and 3.2% dementia from other causes. The key analysis considered the relationship between dementia development and average glucose levels in the preceding 5 years. Those without diabetes (higher glucose level (HGL) 115 mg/dL compared with 100 mg/dL) demonstrated an adjusted HR for dementia of 1.18 (95% CI 1.04 to 1.33). In those with diabetes, HGLs (190 vs 160 mg/dL) were also related to an increased risk of dementia with an adjusted HR of 1.40 (95% CI 1.12 to 1.76).


This study is built on previous work examining ageing, glucose regulation and cognition. Previous research has suggested that diabetes is related to dementia; however, it has suffered from methodological difficulties and has been limited in proposing underlying mechanisms.1 Crane and colleagues address the methodological issues with a large-scale longitudinal study involving multiple measures and a sensitive index of glucose regulation. Their study evaluates a range of blood glucose values and dementia risk in both healthy and individuals with diabetes with precision.

This study did not demonstrate an association between diabetes and dementia; a higher proportion of patients developing dementia did not have diabetes. However, there was an association of increased risk of developing diabetes with increased blood glucose levels in both patients with non- diabetes and diabetes. Although the mechanisms driving the findings are unclear, the authors suggest that prolonged hyperglycaemia may cause vasculature damage in the brain, which may increase the risk of vascular dementia and AD. Previous work has found an association to be stronger between diabetes and AD, when AD is accompanied by cardiovascular diseases.3

In terms of interventions it is worth commenting on other cross-sectional studies that concur with the current findings. High-baseline blood glucose has been found to be a significant predictor of amnestic mild cognitive impairment (MCI) versus ‘normal’ ageing status.4 Therefore, using glucose regulation as a biomarker may aid in the development of both dementia proper and MCI; so as well as being useful in predicting the risk of dementia, an elevated glucose potentially informs strategies for lifestyle and pharmacological interventions to improve glucoregulatory efficiency.

The relationship between efficient glucose regulation and dementia is complex. This article lays the groundwork for three avenues of research: (1) interventions aimed at preventing high blood sugar, (2) the role of glycaemia measures as biomarkers of successful and less successful ageing and (3) the precise neurocognitive mechanism(s) impaired as a result of elevated glucose level.


View Abstract


  • Competing interests None.